The overall goal is to increase our understanding of how thyroid hormone (T3) regulates gene expression. T3 binds to specific receptors (TRs), which bind to T3 response elements (TREs) in specific target genes and thereby regulate transcription. TREs usually consist of two (or more) binding sites ("half-sites") arranged as a direct repeat (DR), palindrome (Pal), or inverted palindrome (IP). TRs can bind to TREs as monomers, homodimers, or heterodimers with retinoid X receptors (RXRs). The relative importance of each of these DNA binding forms in vivo is not known, but it is plausible that each is an important gene regulator under the appropriate circumstance. An hypothesis to be explored is that the "appropriate circumstance" is dictated at least in part by the specific sequence and orientation of the TRE half-sites. The PI has, for example, shown that TR monomers can activate gene expression in transfected cells from a single copy of the 8 bp response element TAAGGTCA, which contains only one half-site. Four Specific Aims will be addressed in this proposal: 1) The TR monomer-DNA interaction will be further characterized. Issues to be addressed include why TR monomers can activate gene expression but RXR monomers cannot, and what portions of the TR molecule dictate binding to the highest affinity DNA binding sequence, TAAGGTCA. Possible differences in the DNA binding specificities of TR-alpha and TR-beta will be explored. 2) Endogenous genes that utilize TR monomer response elements will be identified and characterized. 3) The importance of RXR as a heterodimerization partner will be addressed for differently configured TREs. The hypothesis for this aim is that the sequence and orientation of the response element half-sites dictates the importanCe of RXR. 4) The heterodimerization interfaces of TR will be characterized in the context of TR-RXR binding to differently configured response elements. The hypothesis for this Aim is that the heterodimerization domains of TRs differ on DR, Pal and IP TREs. These experiments derive from preliminary data showing that a potential TR heterodimerization domain known as the ninth heptad plays a different role in heterodimerization on DR, Pal and IP. An additional aspect of this Aim will be to compare the heterodimerization of TR-alpha1 to that of the alternative splice product TR-alpha2, which lacks half of the ninth heptad. The studies with TR- alpha2 also will address the importance of heterodimerization in the ability of this alternative splice product to inhibit T3 action. These 4 Aims will be addressed using a variety of techniques including gel shift, reporter gene expression in yeast and mammalian cells, antisense RNA expression, and mutagenesis. The results should further our understanding of how T3 affects a broad range of metabolic processes in health and disease states such as hyperthyroidism and hypothyroidism. The results also will increase our understanding of the importance of other co- regulators that interact with TRs to generate a T3 response.